TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Show HN: graphiti – Temporal Knowledge Graphs for Agentic Applications

19 pointsby roseway49 months ago
Hey HN -<p>Paul, Preston, and Daniel from Zep here. We’re excited to show you graphiti, a library for building and searching dynamic, temporally aware knowledge graphs.<p><a href="https:&#x2F;&#x2F;git.new&#x2F;graphiti" rel="nofollow">https:&#x2F;&#x2F;git.new&#x2F;graphiti</a><p>With graphiti, you can model complex, evolving relationships between entities over time. graphiti ingests both unstructured and structured data and the resulting graph may be queried using a fusion of time, full-text, semantic, and graph algorithm approaches.<p>With graphiti, you can build LLM applications such as:<p>- Assistants that learn from user interactions, fusing personal knowledge with dynamic data from business systems like CRMs and billing platforms.<p>- Agents that autonomously execute complex tasks, reasoning with state changes from multiple dynamic sources as varied as traffic conditions or streaming voice transcriptions.<p>graphiti differs from GraphRAG and other graph libraries. It’s purpose-built for dynamic data and agentic use:<p>- Smart Graph Updates: Automatically evaluates new entities against the current graph, revising both to reflect the latest context.<p>- Rich Edge Semantics: Generates human-readable, semantic, and full-text searchable representations for edges during graph construction, enabling search and enhancing interpretability.<p>- Temporal Awareness: Extracts and updates time-based edge metadata from input data, enabling reasoning over changing relationships.<p>- Hybrid Search: Offers semantic, BM25, and graph-based search with the ability to fuse results.<p>- Fast: Search results in &lt; 100ms, with latency primarily determined by the 3rd-party embedding API call.<p>- Schema Consistency: Maintains a coherent graph structure by reusing existing schema, preventing unnecessary proliferation of node and edge types.<p>We built graphiti to power Zep Memory, a long-term memory layer for building personalized and accurate LLM apps. We believe graphiti’s potential extends beyond memory applications and have open sourced it to support and grow these use cases.<p>github: <a href="https:&#x2F;&#x2F;git.new&#x2F;graphiti" rel="nofollow">https:&#x2F;&#x2F;git.new&#x2F;graphiti</a><p>documentation: <a href="https:&#x2F;&#x2F;help.getzep.com&#x2F;graphiti">https:&#x2F;&#x2F;help.getzep.com&#x2F;graphiti</a><p>We’d appreciate your feedback, contributions, or just to hear about the awesome projects you’ve built with graphiti!<p>- Paul, Preston, &amp; Daniel

1 comment

hunterbrooks9 months ago
100ms is crazy fast. Approx what&#x27;s the split b&#x2F;t the embeddings calls and the graph traversal work (the library stuff)?
评论 #41380569 未加载